Modelling of Macraes POX Circuit May 2006. Acknowledgements OceanaGold GRD Minproc Brent Hill Tony Frater David King Quenton Johnston Nevin Scagliotta.

Modelling of Macraes POX Circuit

May 2006

Acknowledgements

OceanaGold

GRD Minproc

Brent Hill

Tony Frater

David King

Quenton Johnston

Nevin Scagliotta

Adrian Marin

Presentation Outline

Background

• Macraes POX circuit

• Integration of Reefton concentrates

Modelling

• Metsim model calibration

• Model prediction of increased throughput

Conclusion/Recommendation

Johannesburg Office

Belo Horizonte Office

Macraes Processing Background

Historical Processing

• Small scale operation from 1862 until 1950

• 15 000 oz gold and 100 t scheelite recovered

Modern Processing (Since 1990)

• Crush / Grind / Flotation / CIL

• Crush / Grind / Flotation / Fine grind / CIL

• Crush / Grind / Flotation / Fine grind / POX / CIL

Modern Project History

Major Projects• 1.5 Mt/a sulphide treatment plant – 1990• 3.0 Mt/a expansion – 1994• MREP 4.5 Mt/a - 1999• Increase for sulphide and oxide capacity • Newmont POX technology• 170 t/d BOC cryogenic oxygen plant

Smaller Projects• Unit cell installation• Reclaim circuit• 0.5 Mt/a oxide mill• Autoclave optimisation

Current capacity approximately 6 Mt/a

Macraes Processing Issues 1

• Massive sulphide orebody hosting FeS2 / FeAsS

• Muscovite / quartz/ chlorite / siderite in gangue

• Presence of organic carbon, double refractory

• Variability. Low and high preg-robbing ore types

• 50% to 80% CIL recovery without POX

• Poor recovery with “conventional” POX

Macraes Processing Issues 2

• Newmont technology required for “controlled” POX

• Limestone for free acid control

• Washing for chlorides

• Scale formation in autoclave

Macraes POX Circuit Design

• Concentrate grade 8 - 12 % S

• 3.5 m dia. x 12.6 m

• 2:1 semi-elliptical ends

• 4 agitator, 3 compartment vessel

• 225°C and 3,140 kPag

• Koch Pyroflex membrane and AP302

Autoclave

Scaled Agitator

Reefton Processing

Orebody

• Native gold with minor sulphides in quartz veins

• Gold in FeS2, FeAsS, Sb2S3

Processing

• Crush / Grind / Flotation / Filtration / Transport

• Concentrate at 17.1 % S

• No organic carbon

• Highly refractory, complete oxidation required

Reefton / Macraes Integration

• Additional S oxidation requirement

• Oxygen plant constraint

• Autoclave retention time constraint

• Differing POX conditions

• Requirement for modelling to optimise capacity

History of Macraes POX Modelling

• Spreadsheet POX model developed and verified

• Single-compartment Metsim model developed

• Three-compartment Metsim model developed

• POX chemistry modified based on XRD results

• Thermodynamic data sources consolidated

Plant Trials and Model Calibration

• Plant trial in March 04 generated 23 data sets

• Solids and solution assays recorded

• Operating conditions recorded:

• Autoclave Pressure

• Temperatures in C1, C2 and C3

• Cooling water to C1 ,C2 and C3• Oxygen flow rate and purity

• Overall oxidation from feed and discharge assays

• Compartment oxidation inferred from heat balance

Sulphur Analysis Discrepancy

• Trial data:for 98% oxidation, 20 t/h CW added

• Model results:for 98% oxidation, 16 t/h CW added

• Site assay 10% of the total S (TS) is sulphate S

• No TS reported for the trial data

• No free acid in discharge reported

• Can not do overall S balance calculation

MLA Mineralogy Investigation

• MLA used for quantitative mineralogy investigation

• MLA results 2% of TS is sulfate S

• Site assay 15% sulfate S for the same sample

• Revised S and gangue mineralogy according to MLA

Plant Trials in 10/04 and 01/05

• Updated trial data collection template

• Additional data for heat/mass balance

• Updated mineralogy data used

• Good correlation between models and assays

• No heat adjustment factor required

Plant Trials in 10/04 and 01/05

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

1.00

0 5 10 15 20 25 30 35 40 45 50

RT, min

S2-

oxi

dat

ion

21/10/04 data set1

21/10/04 data set2

28/01/05 data set

Plant High Throughput Trials in 07/05

• In July 2005 eight plant trials run

• Four data sets from scaled autoclave and

• Four sets from “clean” autoclave

• Scaled agitators show poorer oxygen dispersion

• Scaled sets average oxygen utilisation is 79%

• “Clean” sets average oxygen utilisation is 85%

Plant Trials in 07/05

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

1.00

0 5 10 15 20 25 30 35 40 45 50

RT, min

S2-

oxi

dat

ion

Set 2

Set 3

Set 4

Set 5

Set 6

Set 7

Set 1

Set 8

Model to Predict Various Scenarios

• Plot leach kinetics for all plant trials

• Use average kinetic curve for further modelling

• Scenarios modelled:

• Grade: 10%, 12% and 14% total S

• Throughput: 2.7, 2.8, 2.9, 3.0, 3.1 and 3.3 t/h TS

• Constant oxygen partial pressure

• Oxygen: 7 t/h

The Final Kinetic Curve Used for Scenario Modelling

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

0.50

0.55

0.60

0.65

0.70

0.75

0.80

0.85

0.90

0.95

1.00

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

RT, min

% o

f S2-

oxi

dat

ion

21/10/04 data set1 21/10/04 data set228/01/05 data setSet 2Set 3Set 4Set 5Set 6Set 7Metsim input

Scenario Modelling Results

• For 10% S and 12% S- C1 temp drops with higher throughput

• For 14% S- C1 maintains 225°C for all scenarios modelled

S throughput, t/h 2.7 2.8 2.9 3.0 3.1 3.3

10% S 225 224 223 220 218 214

12% S 225 225 225 224 220 218

Compartment 1 Temperature

Scenario Modelling ResultsHigh throughput trials DCS data

0

5

10

15

20

25

30

35

40

20/06/2004 19:12 21/06/2004 0:00 21/06/2004 4:48 21/06/2004 9:36 21/06/2004 14:24 21/06/2004 19:12 22/06/2004 0:00 22/06/2004 4:48

Time

Flo

w t

/h

212

214

216

218

220

222

224

226

228

Tem

p o

C

Flow

C1 Temp

Scenario Modelling Results

• Above 2.7 t/h TS, oxygen constrained

• Increasing throughput, decreases RT for ≤ 12% S

• Increasing throughput, increases RT for ≥ 14% S

• For 14% S the RT is over 50 mins

• The autoclave is not constrained by RT at 14%

Conclusions

• Metsim a useful framework for plant optimisation/design

• Careful selection of chemistry and thermodynamic data

• Plant trial data for model calibration

• Modelling can assist in plant optimisation and future design